Clinical Validation of a Novel Musculoskeletal Modeling Framework to Predict Postoperative Sagittal Alignment.

STUDY DESIGN: A retrospective radiographic and biomechanical analysis of 108 thoracolumbar fusion patients from two clinical centers. OBJECTIVE: This study aimed to determine the validity of a computational framework for predicting postoperative patient posture based on preoperative imaging and surgical data in a large clinical sample. SUMMARY OF BACKGROUND DATA: Short-term and long-term studies on thoracolumbar fusion patients have discussed that a preoperative predictive model would benefit surgical planning and improve patient outcomes. Clinical studies have shown that postoperative alignment changes at the pelvis and intact spine levels may negatively affect postural balance and quality of life. However, it remains challenging to predict such changes preoperatively because of confounding surgical and patient factors. MATERIALS AND METHODS: Patient-specific musculoskeletal models incorporated weight, height, body mass index, age, pathology-associated muscle strength, preoperative sagittal alignment, and surgical treatment details. The sagittal alignment parameters predicted by the simulations were compared with those observed radiographically at a minimum of three months after surgery. RESULTS: Pearson correlation coefficients ranged from r=0.86 to 0.95, and mean errors ranged from 4.1° to 5.6°. The predictive accuracies for postoperative spinopelvic malalignment (pelvic incidence minus lumbar lordosis>10°) and sagittal imbalance parameters (TPA>14°, T9PA>7.4°, or LPA>7.2°) were between 81% and 94%. Patients treated with long fusion (greater than five segments) had relatively lower prediction errors for lumbar lordosis and spinopelvic mismatch than those in the local and short groups. CONCLUSIONS: The overall model performance with long constructs was superior to those of the local (one to two segments) and short (three to four segments) fusion cases. The clinical framework is a promising tool in development to enhance clinical judgment and to help design treatment strategies for predictable surgical outcomes. LEVEL OF EVIDENCE: 3.

Impact of patient, surgical, and implant design factors on predicted tray-bone interface micromotions in cementless total knee arthroplasty.

Micromotion magnitudes exceeding 150 µm may prevent bone formation and limit fixation after cementless total knee arthroplasty (TKA). Many factors influence the tray-bone interface micromotion but the critical parameters and sensitivities are less clear. In this study, we assessed the impacts of surgical (tray alignment, tibial coverage, and resection surface preparation), patient (bone properties and tibiofemoral kinematics), and implant design (tray feature and surface friction) factors on tray-bone interface micromotions during a series of activities of daily living. Micromotion was estimated via three previously validated implant-bone finite element models and tested under gait, deep knee bending, and stair descent loads. Overall, the average micromotion across the tray-bone cementless contact interface ranged from 9.3 to 111.4 µm, and peak micromotion was consistently found along the anterior tray edge. Maximizing tibial coverage above a properly sized tibial tray (an average of 12.3% additional area) had minimal impact on micromotion. A 1 mm anterior tray alignment change reduced the average micromotion by an average of 16.1%. Two-degree tibial angular resection errors reduced the area for bone ingrowth up to 48.1%. Differences on average micromotion from ±25% changes in bone moduli were up to 75.5%. A more posterior tibiofemoral contact due to additional 100 N posterior force resulted in an average of 79.3% increase on average micromotion. Overall, careful surgical technique, patient selection, and controlling kinematics through articular design all contribute meaningfully to minimizing micromotion in cementless TKA, with centralizing the load transfer to minimize the resulting moment at the anterior tray perimeter a consistent theme.

Impact of surgical alignment, tray material, PCL condition, and patient anatomy on tibial strains after TKA.

Bone remodeling after total knee arthroplasty is regulated by the changes in strain energy density (SED), however, the critical parameters influencing post-operative SED distributions are not fully understood. This study aimed to investigate the impact of surgical alignment, tray material properties, posterior cruciate ligament (PCL) balance, tray posterior slope, and patient anatomy on SED distributions in the proximal tibia. Finite element models of two tibiae (different anatomy) with configurations of two implant materials, two surgical alignments, two posterior slopes, and two PCL conditions were developed. The models were tested under the peak loading conditions during gait, deep knee bending, and stair descent. For each configuration, the contact forces and locations and soft-tissue loads of interest were taken into consideration. SED in the proximal tibia was predicted and the changes in strain distributions were compared for each of the factors studied. Tibial anatomy had the most impact on the proximal bone SED distributions, followed by PCL balancing, surgical alignment, and posterior slope. In addition, the thickness of the remaining cortical wall after implantation was also a significant consideration when evaluating tibial anatomy. The resulting SED changes for alignment, posterior slope, and PCL factors were mainly due to the differences in joint and soft-tissue loading conditions. A lower modulus tray material did result in changes in the post-operative strain state, however, these were almost negligible compared to that seen for the other factors.

Validation and sensitivity of model-predicted proximal tibial displacement and tray micromotion in cementless total knee arthroplasty under physiological loading conditions.

The initial fixation of cementless tibial trays after total knee arthroplasty is crucial to bony ingrowth onto the porous surface of the implants, as micromotion magnitudes exceeding 150 µm may inhibit bone formations and limit fixation. Experimental measurement of the interface micromotions is still very challenging. Thus, previous studies investigated micromotions at the bone-tray interface via finite element methods, but few performed direct validation via in vitro cadaveric testing under physiological loading conditions. Additionally, previous models were validated by solely considering relative displacements of the marker couples placed around the tray-bone interface. In this paper, we present an experimental-computational validation framework for investigating micromotions at the tray-bone interface under physiological conditions. Three cadaveric specimens were implanted with cementless rotating-platform implants and tested under gait, deep knee bending, and stair descent loads. Corresponding subject-specific finite element models were developed and used to predict the marker (tray-bone) relative displacements and tibial surface displacements. Experimental measurements were used to validate model estimations. Subsequent sensitivity analyses were performed on implantation and friction parameters to represent model uncertainties. The models appropriately differentiated between locations, activities, and specimens. The average root-mean-square (RMS) differences and correlations between measured marker relative displacements and predictions from the 'best-matching' models were 13.1 µm and 0.86. RMS differences and correlations between measured surface displacements and predictions were 78.9 µm and 0.84. Full-field interface micromotions were investigated and compared with predicted marker relative displacements. The marker relative displacements underestimated the actual interface micromotions. Initial tray-bone alignment in anterior-posterior, flexion-extension, and varus-valgus degrees of freedom have a considerable impact on the interface micromotions. The validated cadaveric models can be further used for pre-clinical assessments of new TKR tray design. The outcomes of the sensitivity analyses provide further insights into reducing interface micromotions via clinical techniques.

Sensitivity of muscle and intervertebral disc force computations to variations in muscle attachment sites.

The current paper aims at assessing the sensitivity of muscle and intervertebral disc force computations against potential errors in modeling muscle attachment sites. We perturbed each attachment location in a complete and coherent musculoskeletal model of the human spine and quantified the changes in muscle and disc forces during standing upright, flexion, lateral bending, and axial rotation of the trunk. Although the majority of the muscles caused minor changes (less than 5%) in the disc forces, certain muscle groups, for example, quadratus lumborum, altered the shear and compressive forces as high as 353% and 17%, respectively. Furthermore, percent changes were higher in the shear forces than in the compressive forces. Our analyses identified certain muscles in the rib cage (intercostales interni and intercostales externi) and lumbar spine (quadratus lumborum and longissimus thoracis) as being more influential for computing muscle and disc forces. Furthermore, the disc forces at the L4/L5 joint were the most sensitive against muscle attachment sites, followed by T6/T7 and T12/L1 joints. Presented findings suggest that modeling muscle attachment sites based on solely anatomical illustrations might lead to erroneous evaluation of internal forces and promote using anatomical datasets where these locations were accurately measured. When developing a personalized model of the spine, certain care should also be paid especially for the muscles indicated in this work.

Twente Spine Model: A thorough investigation of the spinal loads in a complete and coherent musculoskeletal model of the human spine.

Although in vivospinal loads have been previously measured, existing data are limited to certain lumbar and thoracic levels. A detailed investigation of spinal loads would assist with injury prevention and implant design but is unavailable. In this study, we developed a complete and coherent musculoskeletal model of the entire human spine and studied the intervertebral disc compression forces for physiological movements on three anatomical planes. This model incorporates the individual vertebrae at the cervical, thoracic, and lumbar regions, a flexible ribcage, and complete muscle anatomy. Intradiscal pressures were estimated from predicted compressive forces, and these were generally in close agreement with previously measured data. We found that compressive forces at the trunk discs increased during trunk lateral bending and axial rotation of the trunk. During flexion, compressive forces increased in the thoracolumbar and lumbar regions and slightly decreased at the middle thoracic discs. In extension, the forces generally decreased at the thoracolumbar and lumbar discs whereas they slightly increased at the upper and middle thoracic discs. Furthermore, similar to a previous biomechanical model of the cervical spine, our model predicted increased compression forces in neck flexion, lateral bending, and axial rotation, and decreased forces in neck extension.

Twente spine model: A complete and coherent dataset for musculo-skeletal modeling of the thoracic and cervical regions of the human spine.

Musculo-skeletal modeling could play a key role in advancing our understanding of the healthy and pathological spine, but the credibility of such models are strictly dependent on the accuracy of the anatomical data incorporated. In this study, we present a complete and coherent musculo-skeletal dataset for the thoracic and cervical regions of the human spine, obtained through detailed dissection of an embalmed male cadaver. We divided the muscles into a number of muscle-tendon elements, digitized their attachments at the bones, and measured morphological muscle parameters. In total, 225 muscle elements were measured over 39 muscles. For every muscle element, we provide the coordinates of its attachments, fiber length, tendon length, sarcomere length, optimal fiber length, pennation angle, mass, and physiological cross-sectional area together with the skeletal geometry of the cadaver. Results were consistent with similar anatomical studies. Furthermore, we report new data for several muscles such as rotatores, multifidus, levatores costarum, spinalis, semispinalis, subcostales, transversus thoracis, and intercostales muscles. This dataset complements our previous study where we presented a consistent dataset for the lumbar region of the spine (Bayoglu et al., 2017). Therefore, when used together, these datasets enable a complete and coherent dataset for the entire spine. The complete dataset will be used to develop a musculo-skeletal model for the entire human spine to study clinical and ergonomic applications.

Twente spine model: A complete and coherent dataset for musculo-skeletal modeling of the lumbar region of the human spine.

Musculo-skeletal modeling can greatly help in understanding normal and pathological functioning of the spine. For such models to produce reliable muscle and joint force estimations, an adequate set of musculo-skeletal data is necessary. In this study, we present a complete and coherent dataset for the lumbar spine, based on medical images and dissection measurements from one embalmed human cadaver. We divided muscles into muscle-tendon elements, digitized their attachments at the bones and measured morphological parameters. In total, we measured 11 muscles from one body side, using 96 elements. For every muscle element, we measured three-dimensional coordinates of its attachments, fiber length, tendon length, sarcomere length, optimal fiber length, pennation angle, mass, and physiological cross-sectional area together with the geometry of the lumbar spine. Results were consistent with other anatomical studies and included new data for the serratus posterior inferior muscle. The dataset presented in this paper enables a complete and coherent musculo-skeletal model for the lumbar spine and will improve the current state-of-the art in predicting spinal loading.

Implementation of boundary conditions in modeling the femur is critical for the evaluation of distal intramedullary nailing.

In previous numerical and experimental studies of the intramedullary nail-implanted human femur several simplifications to model the boundary and loading conditions during pre-clinical testing have been proposed. The distal end of the femur was fixed in the majority of studies dealing with the biomechanics of the lower extremity, be it numerical or experimental, which resulted in obviously non-physiological deflections. Per contra, Speirs et al. (2007) proclaimed physiological deflections as a result of constraining the femur in a novel statically determinate fashion in combination with using a complex set of muscle forces. In tandem with this, we have shown that not only the deflections but also the stress and strain predictions turn out to be much lower in magnitude, as a result of using the latter approach. To illustrate the dramatic change in results, we compared these results with those of two other models employing commonly used boundary and loading conditions in retrograde stabilization of a distal diaphyseal fracture. The model used herewith resulted in more realistic femoral cortical strains, lower stresses on both the nail and the screws, as well as such deflections in the overall structure.

The effect of loading in mechanical response predictions of bone lengthening.

Intramedullary (IM) distractor nails have become a viable alternative in bone-distraction operations. Upon stabilization of the fractured/dissected limb via the nail, the resulting construct accommodates the load bearing function of the otherwise healthy limb. In establishing design performance targets for these devices, in vitro test conditions are widely accepted leaving the in vivo conditions aside. However, in vivo device failures due to distraction forces necessitate novel modeling considerations. It is especially important to simulate the loads in limb distraction, as this brings the bone-implant construct to a totally different regime than the hip-joint contact force (Point-Force Model, PFM). In this work we used a simplified approach to incorporate ligament stretching due to limb distraction via self-equilibrating spring elements in a finite-element setting (spring-force model, SFM). We compared the effect of loading type on load transmission paths through the locking pins, for these two distinct loading modes, namely, SFM and PFM. The two modes illustrate entirely different load transfer regimes around the bone/nail interface region. In order to avoid high contact stresses between the nail and the bone segments, it is advisable to keep the osteotomy away from the mid-range between the pin connections. It was also seen for both loading modes that including an additional pin at a load transfer location does not significantly alter the load carried by a single pin (the additional pin rather acts as a geometric stabilizer).